As the second decade of managed care's transformation of the US healthcare system comes to an end, consumers the policy and health professional communities, state and federal regulatory, the media, and health plans have begun to shift their attention increasingly to concerns about quality of care. The goal of our proposed program project grant is to understand how marketplace behavior affects quality and how the increased focus on quality in turn affects the various participants in the health care system. We see this system as an interconnected set of markets. One is the market for health care services. Another is the marker for health insurance plans. Since many individuals purchase their insurance through their employer, a third is the market for labor. Behavior and competition in each of these markers will influence outcomes elsewhere in the health care system. This project is designed to understand how these various markets interrelate and how measures of quality can refocus market forces away from a simple concentration on cost minimization to value maximization. Specifically, we will examine: 1) Decisions by large employers in choosing among health plans, 2) how these decisions influence health plan quality in markers with varying degrees of health plan competition, 3) advertising strategies used by plans to attract various types of enrollees, and 4) the implications of behaviors in these different markets for consumers, with a focus on racial and ethnic minority groups. We draw upon pre-existing, large national surveys but will also use data not previously available for research. This includes information about the health plans offered by more than 60 large employers with beneficiaries in over 250 markets, measures of quality from every HMO that has sought NCQA accreditation, a new, market-specific index of provider overlap among plans, and previously untapped sources of health plan advertising data for a panoramic view of the issues under study. We will also use data from in-depth case studies of selected communities to provide detailed portraits of these behaviors in well-described settings, both to assist in understanding the broad data sets, and to facilitate communications of our findings beyond the research community.